A method and system for regulating resource site planning under time-series production simulation

Abstract

The application relates to the field of power system planning, and particularly discloses a regulation resource site planning method and system under time sequence production simulation, which integrates core technical means such as dynamic adjustment strategies, data fusion processing, prediction deviation index quantification and multi-dimensional analysis optimization, realizes comprehensive performance improvement of the regulation resource site planning of a power grid, and specifically realizes dynamic analysis of time sequence adaptation parameters and adjustment of planning strategies, significantly enhances the adaptability of power system planning to multi-time scale load changes, and provides an efficient and reliable solution for resource optimal allocation under a complex power grid environment.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of power system planning, in particular to a regulation resource site planning method and system under time sequence production simulation. BACKGROUND

[0002] In the field of power system planning, time sequence production simulation simulation model has become a key technical means for evaluating system operation and planning flexibility. The initialization input of all basic data and boundary conditions includes power grid topology, annual time sequence load and wind and light output data, existing power supply parameters, candidate installation location and capacity range of regulation resources, and economic constraints. Subsequently, intelligent optimization algorithms (such as genetic algorithms) generate an initial set of first-generation regulation resource site planning schemes in their decision space. Each scheme will be immediately sent to the core time sequence production simulator for comprehensive verification. By constructing a high-resolution time sequence data model containing load, wind and light output, unit characteristics and network topology, and based on economic dispatch and unit commitment optimization principles, the system operation is simulated. Finally, by analyzing the simulation results, the system operation bottlenecks and flexibility gaps are identified, which provides key basis for power structure optimization, energy storage configuration and network upgrade planning decisions. The regulation resource site planning is carried out to plan the access location of energy storage devices, fast-response units, interruptible loads and demand-side response resources. At the same time, in the time dimension, the regulation demand of different time scales is considered, and the construction time sequence, power capacity and energy capacity of the resources are reasonably planned, so as to realize the matching of regulation resources and actual flexibility gap of power system, and effectively reduce the systematic unknown risk brought by high proportion of new energy intervention.

[0003] For example, the Chinese invention patent with publication number CN117856261A discloses a method and device for calculating the regulation capacity gap of power grid based on time sequence simulation. By establishing the operation model of various conventional power sources, the new energy predicted inclusion balance proportion sequence is determined according to the actual operation demand of the power grid and the new energy output prediction value sequence, and the power grid operation line bundle and regulation capacity gap objective function under the new energy output scene are established. Then, under different output combinations of the power grid, the conventional power sources and new energy output in the output combination are adjusted to obtain the optimal solution that satisfies the power grid operation constraint and the regulation capacity gap objective function as the regulation capacity gap sequence of the power grid.

[0004] For example, the Chinese invention patent with publication number CN117559428A discloses an active distribution network planning method based on timing operation simulation, which includes dividing the planning area, making annual 8760 timing prediction on traditional load and new type load in the planning area, simulating annual 8760 timing output of distributed power according to regional meteorological conditions, establishing distribution network framework structure and equipment model through normalization modeling method, using integrated timing simulation analysis to analyze annual 8760 timing power and energy balance results, power flow calculation results, short-circuit current calculation results, power supply safety level results, power supply reliability results, and reactive power planning results, etc., performing technical and economic analysis of planning scheme according to timing results, and then obtaining active distribution network planning results.

[0005] However, in the process of implementing the embodiments of the present application, it is found that the above-mentioned technology at least has the following technical problems: In the field of power system planning, the existing timing production simulation simulation method mainly depends on static or quasi-static model architecture, and its analysis is usually based on highly aggregated global indicators. Such timing production simulation simulation method is difficult to accurately capture the strong randomness and volatility characteristics brought by high proportion of new energy access to power grid, such as the time and space distribution of wind power and photovoltaic output highly dependent on meteorological conditions, the complex coupling relationship between intermittency and anti-peaking characteristics and load dynamics, and the traditional static model cannot effectively represent such multi-time scale dynamic interaction process. In addition, when using an inaccurate timing production simulation simulation method for simulation and obtaining a regulation resource planning method, it is often out of touch with the actual situation. Based on such timing production simulation method, the regulation resource distribution planning scheme formulated is mismatched with the flexibility demand of actual power grid operation in time, space and capacity, causing resource mismatch and even endangering the safe and stable operation of the system. SUMMARY

[0006] In view of the deficiencies of the prior art, the present application provides a regulation resource distribution planning method and system under timing production simulation, which can effectively solve the problems involved in the background art.

[0007] To achieve the above-mentioned purpose, the present application is realized by the following technical scheme: the first aspect of the present application provides a regulation resource distribution planning method under timing production simulation, comprising: S1. marking the timing production simulation simulation power grid as a target power grid, obtaining the fusion timing data of the target power grid, inputting the timing production simulation simulation model and generating the predicted timing sequence, comparing the predicted timing sequence with the actual timing sequence recorded by the target power grid continuously, and determining whether to adjust the configuration parameters of the timing production simulation simulation model.

[0008] S2. initializing the regulation resource site planning strategy of the target power grid based on the predicted time sequence, performing time sequence production simulation, obtaining and analyzing the time sequence adaptation parameters of the time sequence production simulation, and dynamically adjusting the regulation resource site planning strategy of the target power grid.

[0009] S3. recording and analyzing the time sequence adaptation parameter analysis results of the time sequence production simulation, analyzing the operation abnormality type of the target power grid, and determining whether to reconstruct the regulation resource site planning strategy of the target power grid.

[0010] The second aspect of the present application provides a regulation resource site planning system under time sequence production simulation, comprising: a model configuration module, configured to mark the time sequence production simulation model as a target power grid, obtain the fusion time sequence data of the target power grid, input the time sequence production simulation model and generate a predicted time sequence, compare the predicted time sequence with the actual time sequence recorded by the target power grid, and determine whether to adjust the configuration parameters of the time sequence production simulation model.

[0011] A resource planning module is configured to initialize the regulation resource site planning strategy of the target power grid based on the predicted time sequence, perform time sequence production simulation, obtain and analyze the time sequence adaptation parameters of the time sequence production simulation, and dynamically adjust the regulation resource site planning strategy of the target power grid.

[0012] A strategy reconstruction module is configured to record and analyze the time sequence adaptation parameter analysis results of the time sequence production simulation, analyze the operation abnormality type of the target power grid, and determine whether to reconstruct the regulation resource site planning strategy of the target power grid.

[0013] Compared with the prior art, the embodiments of the present application have at least the following advantages or beneficial effects:

[0014] (1) The present application provides a regulation resource site planning method and system under time sequence production simulation, which integrates dynamic adjustment strategy, data fusion processing, prediction deviation index quantification and multi-dimensional analysis optimization, etc. core technical means, realizes the comprehensive performance improvement of the power grid regulation resource site planning, specifically through dynamic analysis of time sequence adaptation parameters and adjustment of planning strategy, significantly enhances the adaptability of power system planning to multi-time scale load change, and provides an efficient and reliable solution for resource optimization configuration in complex power grid environment.

[0015] (2) The present application processes the original time sequence data of the target power grid power supply side, load side, power grid side and energy storage side through the data fusion platform, eliminates abnormal data and fuses meteorological data, forms high-quality time sequence data with spatial tags. This process improves the integrity and accuracy of the data, provides a reliable data basis for the time sequence production simulation model, thereby enhancing the prediction ability and reducing the planning error.

[0016] (3) The present application accurately determines whether the model configuration parameters need to be adjusted by quantifying the cumulative net load fluctuation deviation value, the cross-section flow fluctuation deviation value and the voltage fluctuation deviation value of the predicted time sequence and the actual time sequence. The technology realizes the precision of model configuration, avoids the prediction error caused by improper configuration, and reduces the economic cost and risk of planning failure.

[0017] (4) The present application initializes the adjustment resource distribution planning strategy based on the predicted time sequence, and identifies the key nodes from the space-time hot spot distribution of the line congestion sequence, and combines the congestion duration and the adjustment resource demand for distribution. This method realizes the optimization of distribution through space-time multi-dimensional analysis, improves the adjustment capacity and operation efficiency of the power grid, and significantly reduces the operation cost of the power grid. BRIEF DESCRIPTION OF DRAWINGS

[0018] The present application will be further described with the help of the accompanying drawings, but the embodiments in the drawings do not constitute any limitation on the present application. For ordinary skilled in the art, other drawings can be obtained without creative labor on the basis of the following drawings.

[0019] Figure 1 The present application is a method step flowchart.

[0020] Figure 2 The present application is a system module connection diagram.

[0021] Figure 3 The present application is a target power grid time sequence data processing and model parameter adjustment decision flowchart.

[0022] Figure 4 The present application is a time sequence production simulation model parameter self-adaptive adjustment flowchart.

[0023] Figure 5 The present application is a regulation resource distribution planning strategy dynamic evaluation and adjustment flowchart.

[0024] Figure 6 The present application is a regulation resource distribution planning abnormality processing and strategy reconstruction determination flowchart. DETAILED DESCRIPTION

[0025] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0026] REFERENCE Figure 1As shown, the first aspect of the application provides a method for adjusting resource site planning under time sequence production simulation, comprising: S1. Marking the power grid simulated by the time sequence production simulation as a target power grid, obtaining the fusion time sequence data of the target power grid, inputting the time sequence production simulation model and generating the predicted time sequence, comparing the predicted time sequence with the actual time sequence recorded by the target power grid continuously, and determining whether to adjust the configuration parameters of the time sequence production simulation model.

[0027] S2. Based on the predicted time sequence, initialize the adjustment resource site planning strategy of the target power grid and perform time sequence production simulation, obtain and analyze the time sequence adaptation parameters of the time sequence production simulation, and dynamically adjust the adjustment resource site planning strategy of the target power grid.

[0028] S3. Record and analyze the time sequence adaptation parameter analysis results of the time sequence production simulation, and analyze the operation abnormal type of the target power grid, and determine whether to reconstruct the adjustment resource site planning strategy of the target power grid.

[0029] Specifically, the fusion time sequence data of the target power grid is obtained, and the specific acquisition process is: the fusion time sequence data of the target power grid is obtained through the original time sequence data of the power supply side of the target power grid, the original time sequence data of the load side of the target power grid, the original time sequence data of the power grid side of the target power grid and the original time sequence data of the energy storage side of the target power grid.

[0030] It should be explained that the original time sequence data refers to the key operation parameters recorded in time sequence from the power supply side, load side, power grid side and energy storage side of the target power grid, including load, current and voltage, which are collected through sensors deployed on each side of the target power grid, reflecting the real-time operation state of the target power grid.

[0031] S11. The original time sequence data of the power supply side of the target power grid, the original time sequence data of the load side of the target power grid, the original time sequence data of the power grid side of the target power grid and the original time sequence data of the energy storage side of the target power grid are summarized to the data fusion platform through data transmission.

[0032] S12. The data fusion platform automatically identifies and eliminates abnormal data, and performs timestamp alignment processing on the original time sequence data corresponding to the power supply side of the target power grid, the load side of the target power grid, the power grid side of the target power grid and the energy storage side of the target power grid. At the same time, the short-time missing of the original time sequence data is filled, and the long-time missing of the original time sequence data is completed with historical data of the same period, so as to obtain the initial processing data corresponding to the power supply side of the target power grid, the load side of the target power grid, the power grid side of the target power grid and the energy storage side of the target power grid.

[0033] S13. The initial processing data corresponding to the target power grid power supply side, the target power grid load side, the target power grid grid side and the target power grid energy storage side are integrated according to a unified time axis to generate a space-time matrix, add power grid topology information to form time series data with spatial labels, and fuse meteorological data to enhance the dynamic environmental response characteristics and the space-time correlation characteristics of the target power grid, thereby obtaining the fusion time series data of the target power grid.

[0034] The above data fusion platform refers to a preprocessing platform for power system multi-source heterogeneous data in a time series production simulation simulation model, which integrates intelligent data cleaning, timestamp alignment, multi-scale missing data repair and space-time matrix construction capability based on a unified time axis. At the same time, by deeply integrating power grid topology information and meteorological data, spatial attributes and environmental response characteristics are injected into the original time series data, and finally a high-quality fusion time series data set with spatial labels is generated, which provides a reliable data basis for subsequent high-precision simulation and optimization decision-making.

[0035] It needs to be explained that the short-time missing of the original time series data is filled in by weighting and averaging the adjacent time series data points with the preset short-time missing filling weight coefficient in the power grid planning database to obtain the filling value, and the short-time missing of the original time series data is filled in; it needs to be explained that the long-time missing of the original data is completed by using the historical same period data, and the filling value is obtained by searching the historical same period data in the power grid planning database with the preset long-time missing filling weight coefficient in the power grid planning database for space-time weighting, and the long-time missing of the original time series data is filled in.

[0036] In one specific embodiment, the original time series data of the power grid power supply side, the load side, the power grid side and the energy storage side is first synchronously collected by the terminal, transmitted to the data fusion platform in real time through the power special communication network, and then based on the off-point detection algorithm, the abnormal data including physical limit mutation value (such as load instantaneous zero), sensor constant dead value (long-time invalid reading), communication interruption outlier (obviously deviating from the historical rule) is automatically identified and removed, and then the time series data of the power supply side, the load side, the power grid side and the energy storage side is executed sub-second time stamp alignment to ensure the time consistency of the time series data of each side, and the short-time missing of the original time series data is filled, and the long-time missing of the original data is completed by the historical synchronous data; the preprocessed data enters the space-time fusion stage: based on the unified time axis (such as resolution 15 seconds), the multi-source preprocessed data is integrated, and the dimension usually includes: time step, power grid node, measured physical quantity (such as active power, voltage), and spatial coordinates (such as longitude and latitude), time series, electrical properties and geographical information are integrated into one, providing a complete data base for subsequent analysis, the preprocessed data is associated and mapped with the power grid GIS topology, according to the electrical connection relationship of power plants, substations, lines and other equipment, the spatial coordinates and topological connection properties of the physical node to which each data point belongs are given, so that the originally isolated data is constructed into space-time data with clear spatial position and electrical association; high space-time resolution meteorological data (wind speed, illumination, temperature and humidity, etc.) is synchronously accessed to quantify the nonlinear influence of environmental factors on load and output; finally, the data fusion platform generates fusion time series data containing the state of the power grid body, the spatial topological association and the external environmental driving factor, provides high-fidelity input samples for the time series production simulation model, and supports the dynamic optimization of the regulation resource distribution planning strategy.

[0037] Referring to Figure 3 As shown, first, the fusion time series data of the target power grid is acquired, and then the predicted time series sequence is generated and compared with the actual time series sequence; then the prediction deviation index PI of the time series sequence comparison is calculated, and by judging whether PI is greater than or equal to the prediction deviation threshold, it is finally determined whether to adjust the configuration parameters of the time series production simulation model, thereby forming a complete decision-making process from data input, sequence comparison, deviation quantification to parameter adjustment determination.

[0038] Specifically, whether to adjust the configuration parameters of the time series production simulation model is determined, and the specific determination process is: acquiring the power grid fluctuation parameters of the predicted time series sequence and the power grid fluctuation parameters of the actual time series sequence;

[0039] The power grid fluctuation parameters of the predicted time series sequence include the net load fluctuation of the predicted time series sequence, the section flow fluctuation of the predicted time series sequence and the voltage fluctuation of the predicted time series sequence.

[0040] The net load fluctuation of the predicted time sequence refers to the difference between the predicted total power consumption load and the predicted power generation output of the target power grid; the minute-level difference sequence of the total load and the power generation capacity output by the time sequence production simulation model can be obtained; the cross-section power flow fluctuation of the predicted time sequence refers to the predicted current value of the cross-section of the key position of the target power grid; the time sequence trajectory of the cross-section power flow output by the time sequence production simulation model can be obtained; and the voltage fluctuation of the predicted time sequence refers to the difference between the voltage of the target power grid changing with time and the calibrated voltage; the difference between the voltage of the target power grid changing with time and the calibrated voltage output by the time sequence production simulation model can be obtained.

[0041] The power grid fluctuation parameters of the actual time sequence include the net load fluctuation of the actual time sequence, the cross-section power flow fluctuation of the actual time sequence, and the voltage fluctuation of the actual time sequence.

[0042] The net load fluctuation of the actual time sequence refers to the difference between the actual total power consumption load and the actual power generation output; the difference between the actual total power generation output and the actual total power consumption load at the same time cross-section can be obtained by real-time collection and uploading to the power system dispatch center through the synchronous phasor measurement unit and the remote terminal unit and other sensing devices distributed in the power grid; the cross-section power flow fluctuation of the actual time sequence refers to the current value of the key position of the target power grid; the real-time current value of the key position can be obtained by calculating the massive real-time measurement data collected by the synchronous phasor measurement device, wherein the key position is the position where the circuit branches in the target power grid; and the voltage fluctuation of the actual time sequence refers to the difference between the real-time voltage of the target power grid and the calibrated voltage; the absolute value can be directly obtained by calculating the difference between the bus voltage and the set calibrated voltage and then taking the absolute value through the voltage transformer and the synchronous phasor measurement device deployed in the power grid.

[0043] By introducing the interference weight coefficient, the interference influence degree of the prediction deviation index of the time sequence comparison is obtained by quantifying the proportional relationship between the cumulative net load fluctuation deviation value and the cumulative net load fluctuation deviation of the predicted time sequence and the actual time sequence, the proportional relationship between the cumulative cross-section power flow fluctuation deviation value and the cumulative cross-section power flow fluctuation deviation of the predicted time sequence and the actual time sequence, and the proportional relationship between the cumulative voltage fluctuation deviation value and the cumulative voltage fluctuation deviation of the predicted time sequence and the actual time sequence; the interference influence degrees are coupled to obtain the prediction deviation index of the time sequence comparison, and the specific expression is: ;

[0044] In the formula, PI is a prediction deviation index of the time sequence comparison, pld(t) is a net load fluctuation of the predicted time sequence at the tth sequence comparison point, ld(t) is a net load fluctuation of the actual time sequence at the tth sequence comparison point, pfl(t) is a cross-section power flow fluctuation of the predicted time sequence at the tth sequence comparison point, fl(t) is a cross-section power flow fluctuation of the actual time sequence at the tth sequence comparison point, pvi(t) is a voltage fluctuation of the predicted time sequence at the tth sequence comparison point, vi(t) is a voltage fluctuation of the actual time sequence at the tth sequence comparison point, u_ld is a preset limit net load fluctuation cumulative deviation in the power grid planning database, u_fl is a preset limit cross-section power flow fluctuation cumulative deviation in the power grid planning database, u_vi is a preset limit voltage fluctuation cumulative deviation in the power grid planning database, w1 is a preset disturbance weight coefficient corresponding to the net load fluctuation in the power grid planning database, w2 is a preset disturbance weight coefficient corresponding to the cross-section power flow fluctuation in the power grid planning database, w3 is a preset disturbance weight coefficient corresponding to the voltage fluctuation in the power grid planning database, t is a sequence comparison point of the predicted time sequence and the actual time sequence, t = 1, 2, 3,..., T, and T is the total number of sequence comparison points of the predicted time sequence and the actual time sequence.

[0045] The prediction deviation index of the time sequence comparison is used to represent the prediction deviation degree of the predicted time sequence and the actual time sequence.

[0046] The limit net load fluctuation cumulative deviation represents the minimum critical reference value of the net load fluctuation cumulative deviation, the limit cross-section power flow fluctuation cumulative deviation represents the minimum critical reference value of the cross-section power flow fluctuation cumulative deviation, and the limit voltage fluctuation cumulative deviation represents the minimum critical reference value of the voltage fluctuation cumulative deviation.

[0047] The disturbance weight coefficient corresponding to the net load fluctuation represents the interference influence degree of the proportional relationship between the cumulative net load fluctuation deviation value of the predicted time sequence and the actual time sequence and the limit net load fluctuation cumulative deviation on the prediction deviation index of the time sequence comparison, and the value range is (0, 1]; the disturbance weight coefficient corresponding to the cross-section power flow fluctuation represents the interference influence degree of the proportional relationship between the cumulative cross-section power flow fluctuation deviation value of the predicted time sequence and the actual time sequence and the limit cross-section power flow fluctuation cumulative deviation on the prediction deviation index of the time sequence comparison, and the value range is (0, 1]; and the disturbance weight coefficient corresponding to the voltage fluctuation represents the interference influence degree of the proportional relationship between the cumulative voltage fluctuation deviation value of the predicted time sequence and the actual time sequence and the limit voltage fluctuation cumulative deviation on the prediction deviation index of the time sequence comparison, and the value range is (0, 1].

[0048] It needs to be explained that when the predicted net load fluctuation deviates from the actual value (i.e., |pld(t)-ld(t)| increases), it will directly lead to the deviation of power balance calculation of the power grid, and this power imbalance will drive the corresponding change of the power flow on the transmission section, thereby causing the increase of the prediction deviation of the section flow (i.e., |pfl(t)-fl(t)| increases). The increase of the prediction deviation of the section flow means that there is a difference between the power distribution in the power grid and the expectation, which will directly affect the voltage level of the system, and then cause the increase of the voltage prediction deviation (i.e., |pvi(t)-vi(t)| increases). The deviations of the three links are accumulated and amplified in time sequence, and are in proportional relationship with the defined value (u_ld, u_fl, u_vi) of the corresponding fluctuation of the actual sequence, thereby quantifying the relative deviation scale. In summary, the increase of the prediction error of any fluctuation will not only raise the corresponding proportional term, but also drive the increase of the deviation of the subsequent link through physical correlation, and jointly cause the exponential increase of the prediction deviation index of the time sequence comparison.

[0049] The prediction deviation index of the time sequence comparison is compared with the prediction deviation threshold value, and if the prediction deviation index of the time sequence comparison is greater than or equal to the prediction deviation threshold value, it is determined that the configuration parameter of the time sequence production simulation simulation model is adjusted.

[0050] If the prediction deviation index of the time sequence comparison is less than the prediction deviation threshold value, it is determined that the configuration parameter of the time sequence production simulation simulation model is not adjusted.

[0051] The prediction deviation threshold value refers to the defined prediction deviation index in the power grid planning database, which is the critical benchmark maximum value of the prediction deviation index.

[0052] It needs to be explained that the time series production simulation model first uses the fusion time series data of the historical period as the training set, adjusts the model parameters (including encoder weights, LSTM unit parameters, linear prediction component coefficients, etc.) by the back propagation algorithm to minimize the error between the prediction output and the true value, and often uses sliding window method to construct samples during the training process, and introduces regularization, early stopping and other techniques to prevent overfitting and ensure the generalization ability of the model. Subsequently, the trained model is used to generate the predicted time series sequence (including system level features and regulation demand sequence); after actual deployment, the model also re-trains or fine-tunes the parameters by continuously comparing the latest predicted time series sequence with the actual time series sequence, forming a closed-loop optimization mechanism, so that the model can adapt to the dynamic changes of the power grid and continuously improve the prediction accuracy, completing the training of the time series production simulation model; input the fusion time series data of the target power grid, capture the spatio-temporal dependence of the fusion time series data through the sequence data encoder for feature encoding, input the feature encoding into the core prediction module of the time series production simulation model, learn the time series patterns (including trend, periodicity and random fluctuations) in the fusion time series data, model and predict the linear components of the sequence part through difference, autoregression and moving average operations, learn the non-linear and complex sequence part through long short-term memory, and extrapolate to the future time period. Finally, the time series production simulation model outputs the predicted time series sequence.

[0053] Referring to Figure 4 As shown, first, the predicted time series sequence and the actual time series sequence are uniformly established in the same coordinate system with time as the horizontal axis and regulation resource demand as the vertical coordinate axis; then the average peak value and the average phase are compared step by step - in the average peak value comparison link, according to the different situations that the average peak value of the predicted time series sequence is less than, greater than or equal to the average peak value of the actual time series sequence, respectively, the strategy of increasing, decreasing or not adjusting the equivalent inertia of the system is taken; in the average phase comparison link, according to the judgment result that the average phase of the predicted time series sequence is slower than, faster than or coincides with the average phase of the actual time series sequence, the operation of decreasing, increasing or not adjusting the equivalent effect time delay is carried out accordingly, so as to systematically complete the accurate dynamic adjustment of the key configuration parameters of the model.

[0054] Further, the configuration parameters of the time series production simulation model are adjusted, and the specific adjustment process is as follows: the predicted time series sequence and the actual time series sequence are uniformly established in the same coordinate system with time as the horizontal axis and regulation resource demand as the vertical axis.

[0055] It needs to be explained that the predicted time sequence and the actual time sequence are established in the same coordinate system with time as the horizontal axis and the adjustment resource demand as the vertical axis. First, the time horizontal axis is determined based on the high-precision synchronization timestamp, and the horizontal axis covers the earliest time point of the predicted time sequence and the actual time sequence and the latest time point of the predicted time sequence and the actual time sequence. Second, the vertical axis is defined as the adjustment resource demand, and the vertical axis covers the highest peak value of the predicted time sequence, the highest peak value of the actual time sequence, and the maximum adjustment demand. The predicted time sequence is mapped into the coordinate system and drawn with a specific color and line type, and the actual time sequence is mapped into the coordinate system and drawn with another specific color and line type, and the coordinate system is established. Through this coordinate system, the average peak value and the average phase of the predicted time sequence and the actual time sequence can be directly compared, providing a clear data analysis basis for determining model parameter adjustment.

[0056] The average peak value of the predicted time sequence and the average peak value of the actual time sequence are obtained and compared. If the average peak value of the predicted time sequence is less than the average peak value of the actual time sequence, the predicted deviation index based on the time sequence comparison is matched from the power grid planning database to obtain the equivalent inertia increase coefficient of the time sequence production simulation simulation model system, and the equivalent inertia of the time sequence production simulation simulation model system is adjusted.

[0057] The above-mentioned average peak value of the predicted time sequence refers to the arithmetic mean of the peak value of the predicted time sequence curve in the uniformly established coordinate system.

[0058] The above-mentioned average peak value of the actual time sequence refers to the arithmetic mean of the peak value of the actual time sequence curve in the uniformly established coordinate system.

[0059] It needs to be explained that the predicted deviation index based on the time sequence comparison is matched from the power grid planning database to obtain the equivalent inertia increase coefficient of the time sequence production simulation simulation model system, the preset equivalent inertia increase mapping table of the time sequence production simulation simulation model system in the power grid planning database is called, the predicted deviation index based on the time sequence comparison is input and attributed to the corresponding interval of the predicted deviation index based on the time sequence comparison in the equivalent inertia increase mapping table of the time sequence production simulation simulation model system, and the corresponding equivalent inertia increase coefficient of the time sequence production simulation simulation model system is mapped.

[0060] It needs to be explained that the equivalent inertia of the time sequence production simulation simulation model system is adjusted, the equivalent inertia increase coefficient of the time sequence production simulation simulation model system is multiplied by the system equivalent inertia of the time sequence production simulation simulation model, so as to adjust the equivalent inertia of the time sequence production simulation simulation model system, make the average peak value of the predicted time sequence tend to the average peak value of the actual time sequence, reduce the prediction error of the time sequence production simulation simulation model in the adjustment resource demand, and thus improve the prediction accuracy of the time sequence production simulation simulation model.

[0061] If the predicted average peak value of the time sequence is greater than the actual average peak value of the time sequence, the prediction deviation index based on the time sequence comparison matches the time sequence production simulation model system equivalent inertia reduction coefficient from the power grid planning database, and the time sequence production simulation model system equivalent inertia is reduced and adjusted.

[0062] It needs to be explained that the prediction deviation index based on the time sequence comparison matches the time sequence production simulation model system equivalent inertia reduction coefficient from the power grid planning database, calls the preset time sequence production simulation model system equivalent inertia reduction mapping table in the power grid planning database, inputs the prediction deviation index based on the time sequence comparison and attributes it to the corresponding time sequence comparison prediction deviation index interval in the time sequence production simulation model system equivalent inertia reduction mapping table, and maps out the corresponding time sequence production simulation model system equivalent inertia reduction coefficient.

[0063] It needs to be explained that the time sequence production simulation model system equivalent inertia is reduced and adjusted by multiplying the time sequence production simulation model system equivalent inertia reduction coefficient by the system equivalent inertia of the time sequence production simulation model, so as to reduce and adjust the system equivalent inertia of the time sequence production simulation model, make the predicted average peak value of the time sequence tend to the actual average peak value of the time sequence, reduce the prediction error of the time sequence production simulation model in adjusting resource demand, and thus improve the prediction accuracy of the time sequence production simulation model.

[0064] If the predicted average peak value of the time sequence is equal to the actual average peak value of the time sequence, the time sequence production simulation model system equivalent inertia is not adjusted.

[0065] The predicted average phase of the time sequence and the actual average phase of the time sequence are obtained and compared. If the predicted average phase of the time sequence is slower than the actual average phase of the time sequence, the absolute value of the difference between the predicted average phase of the time sequence and the actual average phase of the time sequence, i.e. the absolute value of the difference between the predicted average phase of the time sequence and the actual average phase of the time sequence, is obtained, and the time sequence production simulation model equivalent effect time delay is reduced and adjusted.

[0066] The above-mentioned predicted average phase of the time sequence refers to the arithmetic mean of the phase values corresponding to each peak value of the predicted time sequence curve in the uniformly established coordinate system.

[0067] The above-mentioned actual average phase of the time sequence refers to the arithmetic mean of the phase values corresponding to each peak value of the actual time sequence curve in the uniformly established coordinate system.

[0068] It needs to be explained that the equivalent effect time delay of the time sequence production simulation model system is reduced by the absolute value of the difference between the average phase of the predicted time sequence and the average phase of the actual time sequence, so as to reduce the equivalent effect time delay of the time sequence production simulation model. The reduction of the equivalent effect time delay can compensate for the response lag of the time sequence production simulation model, synchronize the average phase of the predicted time sequence with the average phase of the actual time sequence, eliminate the prediction deviation of the time sequence production simulation model caused by time misplacement, and thus improve the prediction accuracy of the time sequence production simulation model.

[0069] If the average phase of the predicted time sequence is faster than the average phase of the actual time sequence, the absolute value of the difference between the average phase of the predicted time sequence and the average phase of the actual time sequence is obtained, and the equivalent effect time delay of the time sequence production simulation model is increased.

[0070] It needs to be explained that the equivalent effect time delay of the time sequence production simulation model system is increased by the absolute value of the difference between the average phase of the predicted time sequence and the average phase of the actual time sequence, so as to reduce the equivalent effect time delay of the time sequence production simulation model. The reduction of the equivalent effect time delay can compensate for the response lag of the time sequence production simulation model, synchronize the average phase of the predicted time sequence with the average phase of the actual time sequence, eliminate the prediction deviation of the time sequence production simulation model caused by time misplacement, and thus improve the prediction accuracy of the time sequence production simulation model.

[0071] If the average phase of the predicted time sequence coincides with the average phase of the actual time sequence, the equivalent effect time delay of the time sequence production simulation model is not adjusted.

[0072] Specifically, the adjustment resource distribution planning strategy of the target power grid is initialized based on the predicted time sequence, and the specific initialization process is as follows: S21. The predicted time sequence includes a system-level dynamic characteristic sequence and an adjustment resource demand sequence; S22. Identify the spatial and temporal hotspot distribution of line congestion sequence from the predicted time sequence, count each key node whose congestion duration exceeds the preset defined duration, obtain the system-level dynamic characteristics and adjustment resource demand of each key node; S23. According to each key node, the adjustment resource distribution planning is carried out based on the adjustment resource demand corresponding to each key node, the adjustment resource distribution planning strategy of the target power grid is initialized, and the initialized adjustment resource distribution planning strategy of the target power grid is marked as the initial adjustment resource distribution planning strategy.

[0073] It needs to be explained that, based on the prediction time sequence initialization target power grid regulation resource layout planning strategy, after adjusting the configuration parameters of the time sequence production simulation simulation model, if the prediction deviation index of the time sequence comparison is less than the prediction deviation threshold, the fusion time sequence data is input to the time sequence production simulation simulation model to generate the prediction time sequence, based on the prediction time sequence initialization target power grid regulation resource layout planning strategy, the preliminary prediction of the regulation resource demand in the future period.

[0074] The above-mentioned system-level dynamic characteristic sequence refers to the prediction data sequence of the macro operation state and key performance indicators of the target power grid changing with time, reflecting the overall behavior characteristics of the power grid system, which is a parameter sequence for characterizing the system state with spatial positioning data based on time sequence, including power sequence, current sequence and voltage sequence, etc.

[0075] The above-mentioned regulation resource demand sequence refers to the prediction data sequence of the type, capacity and space-time distribution of various regulation resources required to ensure the safe, stable and economic operation of the power grid under the operation state described by the given system-level dynamic characteristic sequence.

[0076] The above-mentioned preset defined duration refers to the critical maximum value of the statistical blocking duration in the prediction time sequence in the power grid planning database.

[0077] It needs to be explained that, after identifying the space-time hot spot distribution of the line blocking sequence, the key nodes whose blocking duration exceeds the preset defined duration are first identified based on the prediction time sequence output by the time sequence production simulation simulation model, and the embedded out-of-limit detection algorithm will identify all the time periods and line positions whose predicted current values exceed the safe and stable operation limit of the line. The continuous occurrence of these out-of-limit events in time sequence constitutes the line blocking sequence; then the system locates the hot spots through space-time clustering analysis: in the spatial dimension, the spatial statistical analysis unit built in the time sequence production simulation simulation model statistics the frequency and average severity (percentage exceeding the limit) of the blocking of each line, and based on the power grid topology relationship, it identifies those key corridors or areas that are adjacent in electrical distance and often appear blocking at the same time or successively, i.e. spatial hot spots; in the time dimension, the time sequence clustering analysis of the time sequence production simulation simulation model will analyze the time distribution law of the blocking events, and identify the typical time period and seasonal cycle where the blocking occurs most frequently, i.e. the time hot spot; finally, the spatial hot spot and the blocking problem are marked in the global geographical location map based on the time hot spot, and the spatial hot spot is separately noted, and the space-time hot spot distribution map is completed, so as to identify the space-time hot spot distribution of the line blocking sequence, and accurately lock the key nodes whose blocking duration exceeds the preset defined duration based on the space-time hot spot distribution map, thereby providing data basis for subsequent targeted initialization of regulation resource layout strategy.

[0078] In one specific embodiment, the power grid is first simulated for a full cycle by a time-series production simulation model to generate a prediction time-series sequence containing a specific time range. The sequence contains detailed system-level dynamic characteristics and regulation resource requirements. The time-series production simulation model then initiates a congestion analysis based on these data: the embedded out-of-limit detection algorithm scans the predicted power flow of all important lines at a fixed time interval as the resolution, identifying serious congestion events that exceed the line operating limits; through spatiotemporal clustering analysis, it is found that the congestion is spatially concentrated in specific transmission corridors and key areas affected by the concentrated access of renewable energy, and temporally appears stably during the load peak period and the load peak period under specific weather conditions. The visualization and report generation module of the time-series production simulation model automatically generates a spatiotemporal hotspot distribution map and, based on this, calculates the average duration of congestion in the key period that exceeds the system's preset defined duration, thereby determining these nodes as key nodes. Finally, the time-series production simulation model calls its resource planning optimization algorithm to extract the maximum regulation resource requirements of these key nodes during the congestion period, based on which the initial regulation resource distribution planning strategy is initialized, preliminarily planning to build a certain scale of centralized energy storage facilities at key locations, and adding dynamic reactive power compensation devices at specific substations, and marking this specific problem, specific location, and specific capacity-based scheme as the initial regulation resource distribution planning strategy, providing a precise and efficient technical basis for subsequent detailed simulation verification and optimization adjustment.

[0079] Referring to Figure 5 As shown, first, time-series production simulation is performed based on the initial regulation resource distribution planning strategy, and time-series adaptation parameters are obtained to calculate the planning effective index EI; then, whether EI is greater than or equal to the planning effective threshold is determined to decide the subsequent path: if EI is greater than or equal to the planning effective threshold, the regulation resource distribution planning strategy of the target power grid is not adjusted, and if EI is less than the planning effective threshold, the regulation resource distribution planning strategy of the target power grid is dynamically adjusted, which includes two key operations of increasing the distribution density and obtaining the time-series production simulation fluctuation type and updating the heterogeneous regulation resource distribution proportion; the adjusted strategy is marked as a new strategy after dynamic adjustment and is put into time-series production simulation again for verification, and a final judgment is made according to the EI calculated in the new round of simulation: if EI is still less than the planning effective threshold, the regulation resource distribution planning strategy of the target power grid is further adjusted, thereby forming a continuous optimization closed loop based on performance evaluation.

[0080] Specifically, the time-series adaptation parameters of the time-series production simulation are obtained and analyzed, and the specific analysis process is as follows: time-series production simulation is performed based on the initial regulation resource distribution planning strategy, and the time-series adaptation parameters of the time-series production simulation are obtained and analyzed.

[0081] The time sequence production simulation analog simulation time sequence adaptation parameter includes a time sequence production simulation analog simulation mismatch duration, a time sequence production simulation analog simulation mismatch frequency, and a time sequence sequence comparison prediction deviation index.

[0082] The time sequence production simulation analog simulation mismatch duration refers to a duration sum of a deviation between an actual net load and a predicted value in the time sequence production simulation analog simulation exceeding a preset load threshold in a power grid planning database; the time sequence production simulation analog simulation model can output a time sequence curve of an actual load and a predicted load, a sliding window algorithm is used to detect a deviation interval, and then a duration of all continuous deviation intervals is counted and accumulated to obtain the time sequence production simulation analog simulation mismatch duration.

[0083] By introducing interference weight coefficients, the interference influence degrees of a ratio relationship between the mismatch duration and the time sequence production simulation analog simulation mismatch duration, a ratio relationship between the mismatch frequency and the time sequence production simulation analog simulation mismatch frequency, and a ratio relationship between a prediction deviation threshold and the time sequence sequence comparison prediction deviation index are quantified to define the planning effective index of the time sequence production simulation analog simulation, the interference influence degrees are coupled, and the planning effective index of the time sequence production simulation analog simulation is obtained, and a specific expression is as follows: ;

[0084] In the formula, EI is the planning effective index of the time sequence production simulation analog simulation, dl is the time sequence production simulation analog simulation mismatch duration, n is the time sequence production simulation analog simulation mismatch frequency, PI is the time sequence sequence comparison prediction deviation index, dl_max is the preset mismatch duration for defining the mismatch duration in the power grid planning database, n_max is the preset mismatch frequency for defining the mismatch frequency in the power grid planning database, PI_min is the preset prediction deviation threshold in the power grid planning database, k1 is the preset interference weight coefficient of the mismatch duration in the power grid planning database, k2 is the preset interference weight coefficient of the mismatch frequency in the power grid planning database, and k3 is the preset interference weight coefficient of the prediction deviation index in the power grid planning database.

[0085] The mismatch duration represents a maximum critical reference value of the mismatch duration, and the mismatch frequency represents a maximum critical reference value of the mismatch frequency.

[0086] The interference weight coefficient corresponding to the mismatch duration indicates the interference influence degree of the proportional relationship between the mismatch duration and the mismatch duration of the time sequence production simulation on the planning effective index of the time sequence production simulation, and the value range is (0, 1]; the interference weight coefficient corresponding to the mismatch frequency indicates the interference influence degree of the proportional relationship between the mismatch frequency and the mismatch frequency of the time sequence production simulation on the planning effective index of the time sequence production simulation, and the value range is (0, 1]; and the interference weight coefficient corresponding to the prediction deviation index indicates the interference influence degree of the proportional relationship between the prediction deviation threshold and the prediction deviation index of the time sequence on the planning effective index of the time sequence production simulation, and the value range is (0, 1].

[0087] It needs to be explained that when the prediction deviation index increases, the overall accuracy of the time sequence prediction decreases, and the difference between the system running state and the simulation result increases, which will directly lead to the increase of the occurrence frequency of power mismatch in simulation; and the increase of the mismatch frequency will prolong the duration of each mismatch event, and the mismatch duration of the time sequence production simulation will often increase. Continuous and frequent mismatch events further expose the defects of the planning scheme in dealing with actual fluctuations, thereby intensifying the mismatch degree of the system as a whole, which in turn will be reflected in a higher prediction deviation index. The deterioration trend of the three is coupled to the calculation of the planning effective index of the time sequence production simulation through mathematical relationship; in summary, the deterioration of any time sequence adaptation parameter will lead to the decrease of the planning effective index of the time sequence production simulation through its own mathematical term and its driving effect on other indicators.

[0088] Further, the dynamic adjustment target power grid regulation resource distribution planning strategy is adjusted, and the specific adjustment process is as follows: if the planning effective index of the time sequence production simulation is greater than or equal to the planning effective threshold, the regulation resource distribution planning of the target power grid is not adjusted.

[0089] If the planning effective index of the time sequence production simulation is less than the planning effective threshold, the planning effective index of the time sequence production simulation is matched from the power grid planning database based on the planning effective index of the time sequence production simulation to obtain a distribution density increasing coefficient, the distribution density of the time sequence production simulation is increased, the fluctuation type of the time sequence production simulation is obtained, the heterogeneous regulation resource distribution proportion of the time sequence simulation is updated, so as to complete the dynamic adjustment of the regulation resource distribution planning strategy of the target power grid, and mark it as the dynamically adjusted regulation resource distribution planning strategy.

[0090] It needs to be explained that the planning effective index of the time sequence production simulation is matched with the point density increasing coefficient from the power grid planning database, the preset time sequence production simulation point density increasing mapping table in the power grid planning database is called, the planning effective index of the time sequence production simulation is input and attributed to the corresponding time sequence production simulation planning effective index interval in the time sequence production simulation model increasing mapping table, and the corresponding point density increasing coefficient is mapped.

[0091] It needs to be explained that the point density of the time sequence production simulation is increased and adjusted, the point density increasing coefficient is multiplied by the point density of the time sequence production simulation, so that the point density of the time sequence production simulation is increased and adjusted, after the point density is increased, the distribution of the adjustment resource in the physical space is closer to the load center and the new energy rich area, the energy transmission path can be shortened, the network loss can be reduced, the supply and demand balance ability of the local power grid can be improved, the adjustment lag or over-adjustment phenomenon caused by the sparse distribution of the resource can be avoided, and the planning effective degree of the time sequence production simulation is improved.

[0092] The above-mentioned fluctuation type of the time sequence production simulation includes power supply side fluctuation, load side fluctuation, power grid side fluctuation and energy storage side fluctuation.

[0093] It needs to be explained that the fluctuation type of the time sequence production simulation is obtained, and the point distribution proportion of the heterogeneous adjustment resource of the time sequence simulation is updated. First, the full-period fluctuation data recorded in the time sequence production simulation process is analyzed for dominant factors, the fluctuation source type (such as power supply side fluctuation, load side fluctuation, power grid side fluctuation or energy storage side fluctuation) is identified, and the frequency, average amplitude and duration of each type of fluctuation are counted. Then, based on the statistical quantity result, the fluctuation intensity of each type of fluctuation (i.e. the product of the average amplitude and the duration of a single fluctuation) is calculated, and then the fluctuation intensity of all fluctuation events of the same type is accumulated to obtain the total intensity contribution of the fluctuation of this type. Then, the value is divided by the sum of the total intensity contributions of all fluctuation types in the system, so as to calculate the contribution weight of this type of fluctuation in the global fluctuation. According to the contribution weight of different fluctuations in the global fluctuation, the weight of different heterogeneous adjustment resources is correspondingly obtained. The point distribution proportion of the corresponding heterogeneous adjustment resource is taken as the weight of the different heterogeneous adjustment resources, so as to update the point distribution proportion of the heterogeneous adjustment resource. If the power supply side fluctuation (such as wind and light output randomness) accounts for a significant proportion, the point distribution proportion of the fast frequency regulation resource (such as battery energy storage) is correspondingly increased. If the load side fluctuation (such as peak load) is prominent, the point distribution proportion of the medium and long-term adjustment resource (such as pumped storage) is increased. If the power grid side fluctuation (such as congestion event) accounts for the main part, the point distribution proportion of the distributed controllable resource (such as distributed energy storage) around the related line is increased, so as to accurately guide the structural configuration of the heterogeneous resource by quantifying the statistical characteristics of the fluctuation type, and realize the dynamic matching of the point distribution strategy and the actual fluctuation demand.

[0094] Based on the dynamically adjusted regulation resource allocation planning strategy, time-series production simulation is performed to obtain the planning effectiveness index of the dynamically adjusted time-series production simulation and compare it with the planning effectiveness threshold. If the planning effectiveness index of the dynamically adjusted time-series production simulation is greater than or equal to the planning effectiveness threshold, the dynamically adjusted regulation resource allocation planning strategy of the target power grid is marked as an effective regulation resource allocation planning strategy of the target power grid.

[0095] If the planning effectiveness index of the time-series production simulation after dynamic adjustment is less than the planning effectiveness threshold, the regulation resource planning strategy of the target power grid will be further adjusted.

[0096] The aforementioned effective planning threshold refers to the pre-defined effective planning index in the power grid planning database, which is the minimum critical benchmark value of the effective planning index.

[0097] Reference Figure 6 As shown, the system first identifies and analyzes abnormal planning periods, then reduces the starting power during these periods and marks them as target starting power. By determining whether the target starting power is less than the rated minimum regulating resource starting power, the system decides whether to use the rated minimum starting power or the target starting power for strategy adjustment. Based on this, the system further adjusts the regulating resource deployment planning strategy of the target power grid. Then, it records and analyzes the timing adaptation parameter results from the timing production simulation. Finally, based on the analysis results, it determines whether the regulating resource deployment planning strategy of the target power grid needs to be reconstructed. If it is determined that the regulating resource deployment planning strategy of the target power grid needs to be reconstructed, then the regulating resource deployment planning strategy of the target power grid is reconstructed. If it is determined that the regulating resource deployment planning strategy of the target power grid does not need to be reconstructed, then the adjustment process of the regulating resource deployment planning strategy of the target power grid is recorded in the power grid planning database, forming a closed-loop management system for anomaly handling and strategy optimization.

[0098] Specifically, the target power grid's regulation resource planning strategy will be further adjusted. The specific adjustment process is as follows: obtain the simulated fluctuation demand and simulated regulation resource planning records from the time-series production simulation, thereby locating and analyzing the abnormal planning periods.

[0099] It should be explained that the process of locating and analyzing abnormal planning periods first involves acquiring complete output data from the time-series production simulation, mainly including simulated fluctuating demand and simulated adjustment resource planning records. Then, the two time series are compared on a unified time axis. By setting a deviation tolerance threshold, continuous periods where the actual adjustment capacity is consistently lower than fluctuating demand (i.e., insufficient adjustment resources) or significantly higher than demand (i.e., resource misallocation and waste) are identified. The algorithm then comprehensively evaluates the severity, duration, and frequency of deviations within these periods, filtering out those periods that pose significant risks to system safety, economy, or reliability. These periods are then precisely located and marked as abnormal planning periods, providing clear time-series targets for subsequent targeted strategy adjustments.

[0100] The planning effective index based on the dynamically adjusted time sequence production simulation is matched with the start power reduction coefficient from the power grid planning database, the start power of the planning abnormal period is reduced, and is marked as the target start power.

[0101] It needs to be explained that the planning effective index based on the dynamically adjusted time sequence production simulation is matched with the start power reduction coefficient from the power grid planning database, the preset time sequence production simulation start power reduction mapping table in the power grid planning database is called, the planning effective index of the dynamically adjusted time sequence production simulation is input and attributed to the corresponding planning effective index interval in the time sequence production simulation start power reduction mapping table, and the corresponding start power reduction coefficient is mapped.

[0102] It needs to be explained that the start power of the planning abnormal period is reduced, the start power reduction coefficient is multiplied based on the dynamically adjusted time sequence production simulation start power, the start power of the planning abnormal period is reduced, and the planning effective degree of the time sequence production simulation is improved.

[0103] The target start power is compared with the rated minimum regulation resource start power, if the target start power is less than the rated minimum regulation resource start power, the rated minimum regulation resource start power is taken as the planning abnormal period start power, so as to complete the further adjustment of the regulation resource distribution planning strategy of the target power grid.

[0104] If the target start power is greater than or equal to the rated minimum regulation resource start power, the target start power is taken as the planning abnormal period start power, so as to complete the further adjustment of the regulation resource distribution planning strategy of the target power grid.

[0105] Specifically, it is determined whether to reconstruct the regulation resource distribution planning strategy of the target power grid, and the specific determination process is: the planning effective index of the further adjusted time sequence production simulation is obtained, and is compared with the planning effective threshold, if the planning effective index of the further adjusted time sequence production simulation is greater than or equal to the planning effective threshold, it is determined that the regulation resource distribution planning strategy of the target power grid is not reconstructed, and the regulation resource distribution planning strategy adjustment process of the target power grid is recorded in the power grid planning database.

[0106] If the number of times that the regulation resource distribution planning strategy adjustment process of the target power grid appears the same planning abnormal period exceeds the preset upper limit number in the power grid planning database, the operation abnormal type of the target power grid is local response abnormality, and it is determined to reconstruct the regulation resource distribution planning strategy of the target power grid.

[0107] If the planning effective index of the time-series production simulation after further adjustment is less than the planning effective threshold, the operation abnormal type of the target power grid is a system planning abnormality, and the adjustment resource distribution planning strategy of the reconstructed target power grid is determined.

[0108] It needs to be explained that the adjustment resource distribution planning strategy of the reconstructed target power grid, if the operation abnormal type of the target power grid is a local response abnormality, the adjustment resource distribution planning strategy of the reconstructed target power grid is determined, then a local response abnormality warning is performed, and a local response reconstruction scheme is executed; if the operation abnormal type of the target power grid is a system planning abnormality, a system planning abnormality warning is performed, and a system planning reconstruction scheme is executed.

[0109] The above-mentioned local response abnormality warning prompts that the adjustment resource corresponding to the planning abnormal period triggering the local response abnormality warning frequently appears in the distribution planning, and marks that there may be problems such as weak local power grid structure, insufficient resource response capability, and concentrated external condition disturbance, which need to be immediately intervened to avoid the spread of local operation risk.

[0110] It needs to be explained that the local response reconstruction scheme first accurately locates the warning signal. The time-series production simulation model will call the power grid planning database and real-time operation data, cross-compare to accurately locate the specific node, line or distributed energy cluster with excessively high abnormal frequency, and deeply analyze the abnormal causes, such as whether it is a specific feeder overload, unit response delay or local reactive power support shortage and other problems. Next, the scheme enters the strategy generation and deployment stage: the automatic decision engine will calculate and generate a set of conservative and agile adjustment instructions according to the pre-set rule library, for example, re-adjust the output setting value of the fast excitation system, static reactive power compensator or energy storage power station deployed in the abnormal area, temporarily modify the active or reactive power distribution priority of the related unit, or call the mutual aid support capability of the adjacent power grid unit in a very short time, real-time track the execution effect of the above measures and the change of the key operation index, confirm whether the abnormal period is effectively eliminated and whether the adjustment resource is restored to stable response. After confirming that the local area operation state is stable and observing for a period of time, the local response reconstruction scheme is completed, and the whole process log will be recorded in detail and archived to the database for optimizing future warning threshold and response strategy.

[0111] The above-mentioned system planning abnormality warning prompts that the current adjustment resource distribution planning strategy has systematic defects in terms of global time-series coordination, economy or reliability, and marks that there may be problems such as insufficient total amount of resources, imbalance of time and space distribution, and mismatch with load evolution trend, which need to start comprehensive evaluation and reconstruction.

[0112] It needs to be explained that the system planning reconstruction scheme is first based on the latest time sequence production simulation results, and the typical day and typical scene of the long-term operation of the power grid are reviewed, and the core driving factors of the low planning effective index are analyzed by using advanced algorithms, such as whether the inertia of the system is significantly missing due to the high proportion of new energy access, the global shortage of peak shaving capacity, the main network transmission capacity cannot meet the future power flow demand, or the collaborative optimization mode of multiple energy types is defective. After diagnosis, the scheme enters the multi-scheme comparison and iteration optimization stage: reselecting and planning the layout and configuration of large-scale energy storage power stations and optimizing cross-provincial and cross-regional power transactions to introduce external resources, and even adapting the target network to eliminate congestion. Each alternative scheme needs to be put into the time sequence production simulation model again for time sequence production simulation, and a global optimal planning strategy that can make the planning effective index stable and meet the standard and has the optimal technical and economic performance is iterated. Finally, the optimized planning scheme after the review and approval will be officially deployed and executed as the new top-level design, thereby systematically improving the resilience, reliability and economy of the target power grid and ensuring that it can meet various operation challenges in the future.

[0113] Referring to Figure 2 The second aspect of the present application provides a regulation resource site planning system under time sequence production simulation, comprising: a model configuration module for marking the time sequence production simulation of the power grid as a target power grid, obtaining the fusion time sequence data of the target power grid, inputting the time sequence production simulation model and generating the predicted time sequence, comparing the predicted time sequence with the actual time sequence recorded by the target power grid, and determining whether to adjust the configuration parameters of the time sequence production simulation model; a resource planning module for initializing the regulation resource site planning strategy of the target power grid based on the predicted time sequence and performing time sequence production simulation, obtaining and analyzing the time sequence adaptation parameters of the time sequence production simulation, and dynamically adjusting the regulation resource site planning strategy of the target power grid; a strategy reconstruction module for recording and analyzing the time sequence adaptation parameter analysis results of the time sequence production simulation, and analyzing the operation abnormal type of the target power grid, and determining whether to reconstruct the regulation resource site planning strategy of the target power grid.

[0114] The model configuration module is connected with the resource planning module, the resource planning module is connected with the strategy reconstruction module, and the power grid planning database is connected with the model configuration module, the resource planning module and the strategy reconstruction module.

[0115] The above power grid planning database is used for storing the parameters of the regulation resource site planning system under time sequence production simulation.

[0116] The above is only an example and description of the structure of the present application, and those skilled in the art can make various modifications or supplements to the described specific embodiments or replace them with similar ways, as long as they do not deviate from the structure of the present application or exceed the scope defined by the present application, which shall belong to the protection scope of the present application.

Claims

1. A method for regulating resource site planning under time-series production simulation, characterized in that, Comprise: S1. Mark the grid of time sequence production simulation as the target grid, obtain the fusion time sequence data of the target grid, input the time sequence production simulation model and generate the predicted time sequence, compare the predicted time sequence with the actual time sequence recorded by the target grid continuously, and determine whether to adjust the configuration parameters of the time sequence production simulation model; The configuration parameters of the time sequence production simulation model are adjusted, and the specific adjustment process is: The predicted time sequence and the actual time sequence are uniformly established in the same coordinate system with time as the horizontal axis and the adjustment resource demand as the vertical axis; Obtain the average peak value of the predicted time sequence and the average peak value of the actual time sequence and compare them, if the average peak value of the predicted time sequence is less than the average peak value of the actual time sequence, then based on the predicted deviation index of the time sequence comparison, the time sequence production simulation model system equivalent inertia increase coefficient is matched from the grid planning database, the time sequence production simulation model system equivalent inertia is increased, and the time sequence production simulation model system equivalent inertia increase coefficient is multiplied by the system equivalent inertia of the time sequence production simulation model; If the average peak value of the predicted time sequence is greater than the average peak value of the actual time sequence, then based on the predicted deviation index of the time sequence comparison, the time sequence production simulation model system equivalent inertia decrease coefficient is matched from the grid planning database, the time sequence production simulation model system equivalent inertia is decreased, and the time sequence production simulation model system equivalent inertia decrease coefficient is multiplied by the system equivalent inertia of the time sequence production simulation model; If the average peak value of the predicted time sequence is equal to the average peak value of the actual time sequence, the time sequence production simulation model system equivalent inertia is not adjusted; Obtain the average phase of the predicted time sequence and the average phase of the actual time sequence and compare them, if the average phase of the predicted time sequence is slower than the average phase of the actual time sequence, then obtain the absolute value of the difference between the average phase of the predicted time sequence and the average phase of the actual time sequence, and decrease the equivalent effect time delay of the time sequence production simulation model, the equivalent effect time delay of the time sequence production simulation model minus the absolute value of the difference between the average phase of the predicted time sequence and the average phase of the actual time sequence; If the average phase of the predicted time sequence is faster than the average phase of the actual time sequence, then obtain the absolute value of the difference between the average phase of the predicted time sequence and the average phase of the actual time sequence, and increase the equivalent effect time delay of the time sequence production simulation model, the equivalent effect time delay of the time sequence production simulation model plus the absolute value of the difference between the average phase of the predicted time sequence and the average phase of the actual time sequence; If the average phase of the predicted time sequence coincides with the average phase of the actual time sequence, the equivalent effect time delay of the time sequence production simulation model is not adjusted; S2. Based on the predicted time sequence, initialize the adjustment resource distribution planning strategy of the target grid and perform time sequence production simulation, obtain and analyze the time sequence adaptation parameters of the time sequence production simulation, and dynamically adjust the adjustment resource distribution planning strategy of the target grid; S3. Record and analyze the timing adaptation parameter analysis results of the timing production simulation emulation, and analyze the operation exception type of the target power grid to determine whether to reconstruct the regulation resource distribution planning strategy of the target power grid. 2.The method of claim 1, wherein: The fusion timing data of the target power grid is obtained through the following specific process: The fusion timing data of the target power grid is obtained through the original timing data of the power supply side of the target power grid, the original timing data of the load side of the target power grid, the original timing data of the grid side of the target power grid, and the original timing data of the energy storage side of the target power grid. S11. The original timing data of the power supply side of the target power grid, the original timing data of the load side of the target power grid, the original timing data of the grid side of the target power grid, and the original timing data of the energy storage side of the target power grid are collected to the data fusion platform through data transmission; S12. The data fusion platform automatically identifies and eliminates abnormal data, and performs timestamp alignment processing on the original timing data of the power supply side, the load side, the grid side, and the energy storage side of the target power grid, fills in the short-time missing of the original timing data, and completes the long-time missing of the original timing data with historical data of the same period, thereby obtaining the initial processing data corresponding to the power supply side, the load side, the grid side, and the energy storage side of the target power grid. S13. The initial processing data corresponding to the power supply side, the load side, the grid side, and the energy storage side of the target power grid are integrated according to a unified time axis to generate a space-time matrix, add grid topology information to form timing data with spatial labels, and fuse meteorological data to enhance the dynamic environmental response characteristics and space-time correlation characteristics of the target power grid, thereby obtaining the fusion timing data of the target power grid. 3.The method of claim 1, wherein: The configuration parameters of the timing production simulation emulation model are determined according to the following specific process: The grid fluctuation parameters of the predicted timing sequence and the actual timing sequence are obtained. The grid fluctuation parameters of the predicted timing sequence include the net load fluctuation of the predicted timing sequence, the section flow fluctuation of the predicted timing sequence, and the voltage fluctuation of the predicted timing sequence. The grid fluctuation parameters of the actual timing sequence include the net load fluctuation of the actual timing sequence, the section flow fluctuation of the actual timing sequence, and the voltage fluctuation of the actual timing sequence. The interference influence degree of the prediction deviation index of the timing sequence comparison is obtained by introducing an interference weight coefficient to quantize the proportional relationship between the cumulative net load fluctuation deviation value and the defined cumulative net load fluctuation deviation of the predicted timing sequence and the actual timing sequence, the proportional relationship between the cumulative section flow fluctuation deviation value and the defined cumulative section flow fluctuation deviation of the predicted timing sequence and the actual timing sequence, and the proportional relationship between the cumulative voltage fluctuation deviation value and the defined cumulative voltage fluctuation deviation of the predicted timing sequence and the actual timing sequence. The prediction deviation index of the timing sequence comparison is used to represent the prediction deviation degree of the predicted timing sequence and the actual timing sequence. The prediction deviation index of the time sequence comparison is compared with a prediction deviation threshold value, and if the prediction deviation index of the time sequence comparison is greater than or equal to the prediction deviation threshold value, it is determined that the configuration parameter of the time sequence production simulation emulation model is adjusted; If the prediction deviation index of the time sequence comparison is less than the prediction deviation threshold value, it is determined that the configuration parameter of the time sequence production simulation emulation model is not adjusted; The prediction deviation threshold value refers to a preset boundary prediction deviation index in the power grid planning database, which is a critical reference maximum value of the prediction deviation index.

4. The method of claim 1, wherein: The specific initialization process of the adjustment resource distribution planning strategy of the target power grid based on the predicted time sequence is as follows: S21. The predicted time sequence includes a system-level dynamic characteristic sequence and an adjustment resource demand sequence; S22. The spatiotemporal hotspot distribution of line congestion sequence is identified from the predicted time sequence, the key nodes with a congestion duration exceeding a preset boundary duration are counted, and the system-level dynamic characteristics and the adjustment resource demand of each key node are obtained; S23. The key nodes are distributed according to the adjustment resource demand corresponding to each key node, the adjustment resource distribution planning strategy of the target power grid is initialized, and the initialized adjustment resource distribution planning strategy of the target power grid is marked as the initial adjustment resource distribution planning strategy.

5. The method of claim 1, wherein: The specific analysis process of the time sequence adaptation parameter of the time sequence production simulation emulation is as follows: The time sequence production simulation emulation is performed based on the initial adjustment resource distribution planning strategy, and the time sequence adaptation parameter of the time sequence production simulation emulation is obtained and analyzed; The time sequence adaptation parameter of the time sequence production simulation emulation includes the mismatch duration of the time sequence production simulation emulation, the mismatch frequency of the time sequence production simulation emulation, and the prediction deviation index of the time sequence comparison; The disturbance influence degree of the planning effective index of the time sequence production simulation emulation is quantified by introducing disturbance weight coefficients to define the proportional relationship between the mismatch duration and the mismatch duration of the time sequence production simulation emulation, the proportional relationship between the mismatch frequency and the mismatch frequency of the time sequence production simulation emulation, and the proportional relationship between the prediction deviation threshold value and the prediction deviation index of the time sequence comparison, the disturbance influence degrees are coupled, and the planning effective index of the time sequence production simulation emulation is obtained.

6. The method of claim 1, wherein: The specific adjustment process of the adjustment resource distribution planning strategy of the target power grid is as follows: If the planning effective index of the time sequence production simulation emulation is greater than or equal to a planning effective threshold value, the adjustment resource distribution planning of the target power grid is not adjusted; If the planning effective index of the time sequence production simulation emulation is less than the planning effective threshold value, a distribution density increasing coefficient is matched from the power grid planning database based on the planning effective index of the time sequence production simulation emulation, the distribution density of the time sequence production simulation emulation is adjusted, the fluctuation type of the time sequence production simulation emulation is obtained, the heterogeneous adjustment resource distribution proportion of the time sequence simulation emulation is updated, and thus the adjustment resource distribution planning strategy of the target power grid is dynamically adjusted and marked as the dynamically adjusted adjustment resource distribution planning strategy; The fluctuation type of the time sequence production simulation emulation includes power source side fluctuation, load side fluctuation, power grid side fluctuation, and energy storage side fluctuation; The planning effective index of the time sequence production simulation after the dynamic adjustment is obtained, and is compared with the planning effective threshold value. If the planning effective index of the time sequence production simulation after the dynamic adjustment is greater than or equal to the planning effective threshold value, the regulation resource distribution planning strategy of the target power grid after the dynamic adjustment is marked as the effective regulation resource distribution planning strategy of the target power grid. If the planning effective index of the time sequence production simulation after the dynamic adjustment is less than the planning effective threshold value, the regulation resource planning strategy of the target power grid is further adjusted. The planning effective threshold value is a preset critical minimum value of the planning effective index in the power grid planning database.

7. The method of claim 6, wherein: The specific adjustment process of the further adjustment of the regulation resource planning strategy of the target power grid is as follows: The simulation fluctuation demand of the time sequence production simulation and the simulation regulation resource planning record are obtained, so as to locate and analyze the planning abnormal period; The starting power reduction coefficient is matched from the power grid planning database based on the planning effective index of the time sequence production simulation after the dynamic adjustment, the starting power of the planning abnormal period is reduced, and is marked as the target starting power; The target starting power is compared with the rated minimum regulation resource starting power. If the target starting power is less than the rated minimum regulation resource starting power, the rated minimum regulation resource starting power is taken as the starting power of the planning abnormal period, so as to complete the further adjustment of the regulation resource distribution planning strategy of the target power grid. If the target starting power is greater than or equal to the rated minimum regulation resource starting power, the target starting power is taken as the starting power of the planning abnormal period, so as to complete the further adjustment of the regulation resource distribution planning strategy of the target power grid. 8.The method of claim 1, wherein: The specific determination process of the determination of whether to reconstruct the regulation resource distribution planning strategy of the target power grid is as follows: The planning effective index of the time sequence production simulation after the further adjustment is obtained, and is compared with the planning effective threshold value. If the planning effective index of the time sequence production simulation after the further adjustment is greater than or equal to the planning effective threshold value, it is determined that the regulation resource distribution planning strategy of the target power grid is not reconstructed, and the adjustment process of the regulation resource distribution planning strategy of the target power grid is recorded in the power grid planning database. If the number of the same planning abnormal period in the recorded adjustment process of the regulation resource distribution planning strategy of the target power grid exceeds the preset upper limit number in the power grid planning database, it is determined that the regulation resource distribution planning strategy of the target power grid is reconstructed. If the planning effective index of the time sequence production simulation after the further adjustment is less than the planning effective threshold value, it is determined that the regulation resource distribution planning strategy of the target power grid is reconstructed.

9. A system for regulating resource deployment planning under time-sequential production simulation, applying the method for regulating resource deployment planning under time-sequential production simulation according to any one of claims 1-8, characterized in that: The model configuration module is configured to mark the time sequence production simulation power grid as a target power grid, obtain the fusion time sequence data of the target power grid, input the time sequence production simulation model and generate the predicted time sequence, compare the predicted time sequence with the actual time sequence recorded by the target power grid, and determine whether to adjust the configuration parameters of the time sequence production simulation model. ​ The resource planning module is configured to predict a timing sequence, initialize an adjustment resource distribution planning strategy of the target power grid, and perform timing production simulation, obtain and analyze timing adaptation parameters of the timing production simulation, and dynamically adjust the adjustment resource distribution planning strategy of the target power grid. The strategy reconstruction module is configured to record and analyze the timing adaptation parameter analysis results of the timing production simulation, analyze the operation abnormal type of the target power grid, and determine whether to reconstruct the adjustment resource distribution planning strategy of the target power grid.

Images